Tuberculosis remains a major public health problem worldwide. The management of tuberculosis has recently become increasingly difficult with the rising incidence of drug resistant disease. The causative organism, Mycobacterium tuberculosis, is innately resistant to many common antibiotics and accumulates mutations that allow resistance to multiple antibiotics. Resistance now necessitates the use of less potent and more toxic second line drugs for treatment of many infected patients. While the mechanisms of drug resistance to several first line agents has been determined, little is known about second line drugs. We propose to investigate the molecular mechanisms of resistance to both antimycobacterial antibiotics and agents not ordinarily used to treat tuberculosis. We will utilize methods for generating and analyzing transposon mutants that we have recently developed. These methods have already identified previously unknown mutations responsible for drug resistance. We will also study the population dynamics of M. tuberculosis during infection in a mouse model. Our mathematical model suggests that drug potency and the physiologic "cost" (effect on growth rate) of resistance are more important in the development of resistance than mutation rate, an observation that would affect the design of antibiotic treatment strategies. We will use "tagged" strains to follow the fate of individual strains, both drug sensitive and drug resistant, during infection. This will allow us to compare the relative fitness of strains and determine the cost of drug resistance in the presence and absence of antibiotic treatment. The results of these studies will be useful for developing new antibiotic treatments and strategies